Turtle Sense

Cell phones for Sea Turtles and beyond -- creating an extremely low powered remote data recorder and sensor for monitoring wildlife, etc...

Similar projects worth following
This very low powered, inexpensive remote data recording system was first designed to monitor the nests of endangered sea turtles, but can be adapted for many other purposes. Designed to withstand a harsh ocean beach environment, an ultra low powered sensor and microprocessor are combined on a tiny circuit board for applications in difficult locations. For turtle monitoring it is encased in a small plastic egg. This "Smart Sensor" - buried in the sand on top of a nest - measures, records and evaluates motion and temperature for several months. The Smart Sensor is attached by cable to a cell phone board which sends regular reports on the activity in the nest. The goal is to predict hatching dates from motion of the embryo in the egg or from motion of the egg as the hatchlings cut through the leathery egg or both. The smart sensor and cell phone board can be easily modified with other sensors or wireless communication.

The sea turtles found in the waters of Cape Hatteras National Seashore are protected by the U.S. Endangered Species Act, an extremely powerful law, that must be followed and administered by the National Park Service (NPS). While the NPS must protect the Seashore’s nesting sea turtles, the NPS also is obligated to uphold it’s mission and make the Seashore available to people for “enjoyment, education, and inspiration.” As the popularity of the Seashore has grown, it has become increasingly difficult for the NPS to meet its obligations to both people and sea turtles. The goal of Turtle Sense is to help people and sea turtles share the beaches of the Seashore in a way that benefits both humans and sea turtles.

Today, when a sea turtle nest is found on a Seashore beach, a small enclosure is built around the nest to keep pedestrians and vehicles away. About 50 to 55 days later the nest closure is expanded, often closing the beach to vehicular traffic. Because there is no reliable way to predict when tiny turtles will emerge from their nests near the duneline and parade to the surf, closures can sometimes last for more than a month. 

Turtle Sense system design

The Smart Sensor (the right half of the diagram above) uses an accelerometer to record changes in acceleration up to 400 times per second.  The readings are analyzed by the microprocessor to produce a profile summary of the forces acting on the sensor.  A new profile is created every 15 seconds to six minutes. Because it is not possible to transmit data from underneath wet salty sand, a set of summaries are uploaded to a separate Communications Unit (the left half of the diagram) at least once a day and as often as every hour.  The Comm Unit and the Smart Sensor are connect by Cat5e Shielded cable, which is typically about 20 feet (6 meters) long, but can be much, much longer if needed.  The Comm unit is controlled by another microprocessor which controls the communication with the Smart Sensor and also controls a plug-in cellular communications board.  Cellular communication uses the bulk of the power needed by the system, so its power supply is powered down when communication is not needed.  It is typically only powered up for a few minutes each day.  The entire device is powered by a battery pack of 8 rechargeable AA NiMH cells.  This is enough power to run the system for many, many months.

In order to keep track of multiple nesting sites, we also created a hand-held communications device that is used to test the sensors, check for good cell phone reception, and register the date and GPS location of the nesting sites.  Since this device is small and portable, it can be carried along with Smart Sensors on the daily beach patrols of park personnel.  When nests are found, they are excavated and the eggs are counted.  At that time, a Smart Sensor is placed on top of the nest, and it is connected to the hand-held device to register the nest.  Testing and registering the sensor just takes a few minutes.  Then the cable is buried, with its 9-pin Molex connector sealed in a plastic pill bottle.  The location of the end of the cable is marked with a stake.

After a few days or weeks, a larger comm unit is brought to the nest site and connected to the sensor.  This Comm unit is housed in sealed PVC pipe in a foundation of concrete. The comm unit enters a sleep mode if no sensor is plugged in. It powers up fully when a Smart Sensor is connected, and starts sending reports.

Detailed reports are regularly uploaded to a server on the Internet.  These text file reports are readable by humans and machines.  Eventually, there will be a web based interface for controlling the parameters of the system and viewing data.  If we are successful, an automated system will accurately predict hatching events and send alerts of pending hatching to wildlife managers, researchers and the general public.  To make that possible, we need to analyze the data correlated with human observations at several nesting sites to...

Read more »

  • 2 × ***COMPLETE BOMs FOR CIRCUIT BOARDS ARE IN THE LIST OF LINKS*** The links are labeled "Custom Smart Sensor BOM" and "Custom Communicator BOM".
  • 1 × Smart sensor custom circuit board (1" X1") Components include TI MSP430FR5739 processor, ADXL 362 motion sensor and ADM 3491 transceiver. See "Custom Smart Sensor BOM" link in left column.
  • 1 × Communications custom circuit board (3.2" X 1.5") components include TI MSP430FR5739 processor, headers for cell phone board, power supplies, analog switches, and ADM 3491 transceiver. See "Custom Communicator BOM" link in left column.
  • 1 × Janus plug-in M2M cell phone boards 5 interchangeable boards that are certified to work with different telecoms around the world -- NOTE! This 3rd party board is not open hardware, but it is FCC approval and carrier certified, which is a difficult, and expensive process. We are not aware of any open hardware, FCC approved and carrier certified cell phone boards.
  • 1 × Cell phone antenna Penta band with right angle SMA connector

View all 48 components

  • 2015 Update

    Samuel Wantman08/28/2015 at 03:51 0 comments

    We've been hard at work on the next phase of this project. We've installed a capacitive moisture detector in the egg sensor, changed from 8 AA NiMH batteries to 3 AAAs, added a solar cell that keeps the batteries charged up until they totally die (about 10 years?), and changed the connection between the sensors and the communications tower to a single RG6 coax cable. The coax will power the sensors and also carry data. A single communications tower will be able to handle multiple sensors simultaneously (we're not sure how many yet, but it should be about a couple dozen). The coax should be much more reliable than the Cat5 cable we have been using, and make it easier and cheaper to build and maintain. A coax cable can be repaired in the field in about a minute. There will be lots of ways to extend the system using SPI, I2C, and analogue and digital connections. Software will be field updatable, and we've been working on server-side visualization of the data. For more information about our progress, please visit

    The National Park Service has been using the system this summer with excellent results. One sensor failed, but every other nest has provided data that was used to accurately predict when baby sea turtles would emerge from their nests. Most of our predictions were about 4-5 days before emergence, and the prediction was usually correct within a day. We were able to spot nests that were infertile, and were only incorrect when we predicted an infertile nest, but in fact less than 10 percent of the eggs hatched. While we missed that handful of hatchlings, the NPS was able to rescue about 75 hatchlings from almost certain drowning because of our system. We knew that they had hatched, but they hadn't yet emerged. A storm and high tide threatened to drown the hatchlings who were trapped in the nest with a hard crust above it. Because the NPS knew from our system that there were hatchlings, they were able to rescue them just in time.

  • Unintended Consequences (mostly positive)

    Samuel Wantman10/02/2014 at 14:23 0 comments

    I just returned from nine days visiting our two test sites in North Carolina. While there, I got a chance to see baby sea turtles making their way to the ocean and to talk with wildlife managers and biologists. I learned quite a bit from the trip. We went out for several nights to visit a nest where I had predicted a boil using the data from our sensor. The turtles emerged two days later than expected, but that was not surprising considering that the temperature had been cooling down in the prior two weeks, and the previous nests had hatched during warmer periods. The boils often occur between sundown and about 10:00 PM, which is when we usually gave up and went home. This nest ended up boiling a little after midnight.

    Read more »

  • User Interface

    Samuel Wantman09/16/2014 at 03:15 0 comments

    We want to make Turtle Sense as simple to use as possible. So, at the most basic level, setting it up, you just plug the connector coming from the turtle egg Smart Sensor into the connector attached to the communications tower. The device, which has been asleep, wakes up and starts collecting data and sending out reports. When you unplug the devices, they go back to sleep until the next time.

    But that is just the surface of how we plan to have users interact with the device. We made these units for use by the National Park Service, but the NPS is just the representative for a much bigger group of users: wildlife managers, researchers, eco-tourists, fisherman, and the general public.

    Read more »

  • Future options

    Samuel Wantman09/16/2014 at 01:22 0 comments

    The 2014 Turtle season will be over in a couple of months. We will have collected data from more than a dozen nests, and have been fairly successful in accomplishing what we set out to do this season. But we still have a long way to go. We need to automate the process of predicting nests and create a fully functional web-site for managing the process that also lets the public and researchers see the results. At the same time we have to perfect our design so that it is as close to 100% reliable as we can make it. Here's our thinking about the next phase of the project.

    Read more »

  • Hatching or Ghost Crabs?

    Samuel Wantman09/12/2014 at 08:45 0 comments

    We've completed the monitoring of five nests so far. One sensor failed in a nest (due to a bad connection). Eight nests are being monitored currently, and another four are awaiting a spare Communications Unit to become available. So, we should have data from about 17 nests to evaluate. Using our theory about hatching, which we are now calling the "popcorn" theory, we were able to predict several boils. The popcorn theory is that the turtle eggs are like popping corn in boiling oil. When the kernels heat up, they start to jiggle a little, and then they all start to pop at once. When you hear the popping quiet down, you know it is time to pour everything out of the pot. So, we theorize that the turtles are all programmed to listen for things to quiet down after hatching, and that way they know that it is time to leave the nest.

    Popping (hatching) shows up in our data a few days (3-5) before the turtles pour (boil) out of the nest. The sensors are sending 240 records in each report, and each report is phoned in every 4 hours. So each record is a profile of what happened each minute of the day. We can see how many readings there were in about 25 ranges of acceleration from .001 G to about 4 G. We've arranged these ranges logarithmically because we did not know what values we'd be getting. We thought there would be a huge dynamic range between background noise and the motion of hatching turtles hitting the sensors. It turns out the dynamic range is not that large--about 4:1. In future versions, we can design our data collection to get more resolution in the ranges that we are recording.

    To graph the data, we integrate all the readings during each minute's report to get a single numerical value. This value corresponds roughly to the energy from the motions recorded. We can graph the data versus time to see what is happening. It looks like this:

    Looking at the graph, you can see that things are really popping starting around day 68. Then everything gets really quiet for a few hours on day 72. That was just before the boil. So, when we see the graph look like an earthquake on a seismograph, we figure things are going to boil in about 4 days. This is a pretty good method of making a prediction.

    Except for one thing. Do you see the spike around day 60 and again on day 65? We think these spikes are from ghost crabs. Ghost crabs are a main predator of sea turtle eggs. They make quite a commotion in the nest when they are feeding on sea turtle eggs. Some nests don't have them, so there are no spikes until the eggs are hatching. But some nests have lots of them. If the crabs find the nest early on, they return over and over. They probably bring their family and friends to feast on eggs.

    So now we have to collect more data and learn how to distinguish between the hatching of sea turtles and an orgy of crabs feasting on sea turtles. One member of our team, Tom Zimmerman, is working on using machine learning techniques to do just that.

  • Real World Problems

    Samuel Wantman08/30/2014 at 23:52 0 comments

    We now have about two dozen sensors in the field and about a dozen communicators. Early on, we made a test unit. That unit has been operating for weeks sitting on a counter here in San Francisco, and it has never skipped a beat, missed a report, or had a problem of any kind since we got everything working. The units in the field have not all been so reliable. Some of them have worked flawlessly for months, and some of them never powered up. Over all, they are working about 90 percent of the time. Testing electronics in the wet, salty sea environment tests your design in extreme ways.

    We are building and programming more than 2500 miles from where the units are being deployed. We need them to run, unattended, for more than two months, enduring moisture, salt, wind, blowing sand, and who knows what else. When they stop working, we can't easily determine what went wrong, and that is a challenge. With two custom devices interconnected and running a third device talking to a server, there are lots of places where things can go wrong. This is where "hacking" turns into full-scale engineering.

    It is one thing to get the basic functions of a device up and running. That part is fun--it requires innovation and creative design. But getting everything to work flawlessly in the field requires diligence, attention to detail, and lots of trouble shooting. That is much less sexy. For this project to be successful, we have to make units that are totally reliable. Ninety percent is not sufficient.

    We've learned quite a bit about the ways things can go wrong. Boards can come loose, connections can break, moisture and salt can get into places they are not supposed to get into. Cell phone reception can be interrupted by lightning, and that can cause embedded code to hang.

    We anticipated several processes that could fail and programmed the code to handle the problems as they happened. For everything else, there is a watchdog timer that resets the device. But it is difficult to test how things will behave when they fail, if you can't observe them failing. We don't get thunderstorms in San Francisco, and being in the middle of the city, the cell phone reception is excellent. This summer is teaching us about all the problems we will need to deal with.

    There are two main things we have learned thus far. First, the watchdog timer works for most unanticipated problems, and the system recovers most of the time. But we need to handle recovering from resets better. Sometimes the system recovers and no data is lost, but sometimes we lose a few hours' or a day's worth of data. We knew this might happen, and we knew how to fix it, but unfortunately, we ran out of code space in the 16K devices we are using (TI's MSP430s) to implement the fixes. Just at the point when we ran out of code space, TI introduced new versions with up to 128K of memory. These new processors should be more than adequate for what we need. But making these fixes is out of the question for this year. The devices are in the field, and the turtle season is half over. So for now, we are accumulating more knowledge about the ways our devices can fail. We will incorporate recovering from those failures into the next version of the software.

    Second, the connection between the Smart Sensor and the Communications Unit is the weak link in our design. We consulted with a few engineers about what connector to use, and we ended up choosing a 9 pin Molex connector with gold contacts. The first thing we discovered was that the cable connection to the connector was prone to failure because the outer cover of the cable was removed near the connectors. In the field, the cable bent easily during connecting and disconnecting. The thin wires in the Cat5e cable could easily break after a few uses. So we cast the ends of the connectors and the cables in quick-setting polyurethane to make it strong. It did make it strong, but the polyurethane wicked its way up into the connector and made some of the crimp connections...

    Read more »

  • Results from the First Nest

    Samuel Wantman08/11/2014 at 10:00 0 comments

    We had our first boil yesterday, August 10th, and I'm happy to report that the day before the boil, I sent an e-mail to the National Park Service saying that I thought something might be happening in the nest. Given this, it might actually be possible to predict when a nest will hatch!

    After the boil, I spent a few hours looking at all the data to see what I could find, and the results were not quite what I had expected. In fact, I have a new hypothesis about how to predict when a boil is about to occur.

    Previously, I had heard that the activity from the turtles moving around in their eggs and then from hatching stimulated the other turtles to hatch, and that somehow they coordinated the boil from all this motion. From looking at the data, I think what might be happening is that the motion ramps up gradually until they start hatching. Then, as they hatch, the motion becomes erratic. Finally, after they all have hatched and are waiting for the boil, everything quiets down. It is this quieting down that might be the clue to the turtles that all their siblings have hatched and it is time for a boil. It makes sense, since as each turtle hatches, there is no need for it to move any more. All it needs to do is wait for everyone else to finish their wiggling. If there is no wiggling, all the eggs have hatched, and it is time to go.

    From looking at the orientation of the sensor, it appears to have rotated a little between 2 PM and 7 PM on August 7th, which appears to be the period when there were the largest jolts. This makes sense if this movement was from the collapse of the nest as the turtles hatched. The sensor would get lower, and its orientation would rotate about as much as can be seen in the data.

    It will be interesting to see what the data looks like from some more hatching nests and whether they all look similar. Who knows? Another nest might be boiling as I write this . . .

    The data was also quite surprising in another respect. The boil itself is hardly noticeable. In fact, I'm not exactly sure what time it occurred from looking at the data. There were no dramatically large motions. I thought for sure that we would get very high readings when the baby turtles moved past the sensor egg. It looks like the sand might have absorbed much of the energy. It might also be that when the nest collapsed, the sensor did not sink with it as much, and very few of the turtles had any contact with the sensor. It will be useful to know the orientation of the sensor relative to the egg shells when nests are excavated.

  • Hurricane? Yes. Postal Service? No.

    Samuel Wantman08/06/2014 at 07:04 0 comments

    So, we got through Hurricane Arthur without a problem, and we decided to send some more Communications Units into the field. We sent them out via the USPS Priority Mail. The units are housed in schedule 40 PVC pipe, which is really, really strong, so we thought that there wouldn't be a problem in the mail. We were wrong. It turns out that we planned for everything except for the abuse that the units would receive in transit.

    Most everything in the device is soldered together, but there are a few connections that are not. The cell phone board is plugged into headers. There are several boards available from the manufacturer that work with different telecoms. The boards could be soldered in, but we wanted to be able to swap in a different board for different carriers if need be. With 49 pins holding it in, it takes quite a bit of effort and wiggling to get it out and seemed very, very secure. But not secure enough for the USPS! Similarly, the eight AA batteries are secured in a battery pack, and the pack can be unplugged with a standard 9 volt battery clip. These take quite a bit of force to put on and take off, but the trip cross country in the mail was enough to shake it loose. None of the units in our last shipment was working by the time they made it from California to North Carolina.

    The repair for these problems is very simple, and in retrospect we might have anticipated this issue. To secure the board and the battery clip, we tightened cable ties around them. There is no way to get them loose without cutting the ties off.

    So the moral of the story is, when you are designing a product, don't worry about making it hurricane proof. Make sure it can make it through the post office.

  • Summer 2014 Field Notes

    Samuel Wantman08/06/2014 at 06:48 0 comments

    We currently have about a dozen sensors installed in nests, but only three of them are connected to communications devices. For the first month after a clutch of eggs is laid, there is little or no activity in the nest. Since this is the first testing of our devices in the field, we want to fix any bugs in the first devices before sending out lots more.

    The first communicators were installed around the end of June, just in time to be tested by some very extreme conditions during Hurricane Arthur. We didn't know what to expect. Would the cement anchors on the PVC pipes that house the communicators be secure enough? Would water find a way into our hermetically sealed enclosures? Would the nests get washed away? We changed the parameters on the units so that they phoned in every two hours instead of once a day. We figured if the units failed, we'd get information right up until two hours before they failed. As we watched the hurricane forecasts, the storm track was pointing directly to where our first devices were installed.

    The units made it through the Category 2 hurricane without any problems. So, now we know that they can handle severe weather. We also learned something very important from the storm. We had been worrying that environmental noises and disturbances might make it hard to tell the difference between those disturbances and hatching eggs. What we found from the data during the hurricane is that it is very quiet a foot or two feet below the beach. The noise level from crashing waves, blowing sand, and torrential rain was barely noticeable in our data. On our logarithmic scale, that means that the noise level during the storm is no more than about 40 percent above normal. Since our scale has about a 1:1,000,000 range, 40 percent is not significant. The noise level we are encountering in the nests is comparable to the noise level on the table in my empty kitchen when I was testing the sensors overnight.

    So what does this mean about the motion we will record as the embryos start moving and later when the eggs start hatching? This is the big unknown. We'll find out soon.

  • Phase Two implementation (2014 Sea Turtle Season)

    Samuel Wantman08/05/2014 at 20:28 0 comments

    Test units started being installed in sea turtle nests in June of 2014. The plan is to monitor about 20 nests at the Cape Hatteras National Seashore and a few more at the Bald Head Island Conservancy, both in North Carolina.

    The units constantly monitor and analyze motion to create a profile of its magnitude over time. The motion detector measures the change in acceleration (or "jolt") multiple times per second. The magnitude of the jolt is placed on a logarithmic scale divided into 25 different ranges. The ratio between the lower and higher limit of each range is the square root of 2. A counter for each of the different ranges is incremented whenever a jolt reading is in that range. You can think of each counter as a bin. If a magnitude is in range, it is thrown into the bin for that range. After 6 minutes the results are stored in a record along with a temperature and orientation reading, and a new set of bin counters is started. By looking at how many readings were in each bin, we can get an idea of what jolts occurred while the bins for that record were being filled. This allows us to compress several thousand readings into approximately 32 bytes of information. We lose precision (including the exact sequence) with all the readings, but we suspect that those details are not important. The 240 records created each day give us a very good idea of what is happening in the nest.

    Research indicates that before emerging from the nest in a "boil," turtle hatchlings congregate underground near the top of the nest. It is thought that this motion stimulates the hatching of the turtles that haven't yet emerged. Our sensors, situated at the top of the nest, should record some large disturbances when the first turtles emerge. It is our hope that our readings will indicate these early hatchings so that we can predict that a boil is imminent.

    The nests with first units installed should be hatching around mid-to-late August. We will be generating algorithms to predict hatching from our data, and we plan to refine those algorithms so that we can come up with a reliable process for predicting hatching a few days in advance.

View all 12 project logs

View all 6 instructions

Enjoy this project?



PointyOintment wrote 03/19/2015 at 03:14 point

You might be interested in #Project Apollo

  Are you sure? yes | no

Samuel Wantman wrote 08/05/2014 at 22:42 point
We are measuring the temperature in the nest. In our situation, we only have official permission to install a single sensor in the nest. Our hope is to use the temperature data with the motion readings to refine our estimations on the total incubation period. It seems that the effect of global warming on the sex of sea turtles is a hot button topic. As soon as our design stabilizes, we will be making units available at cost for researchers around the world. We will also be looking for people to help test our units on GSM networks outside of the US, as we hope to make the device compatible with cell phone networks all over the world. We will also be encouraging researchers to expand on our software so that it is applicable for other uses with sea turtles and with many other egg-laying species.

  Are you sure? yes | no

jlbrian7 wrote 08/09/2014 at 10:45 point
I don't think her work dealt with global warming. Apparently it is not known with certainty what temperature ranges produce males and what produce females. I think it is hard to know what impact climate change has when you don't have a reliable baseline.

  Are you sure? yes | no

jlbrian7 wrote 08/05/2014 at 22:32 point
My wife was awarded a Fulbright to study the arribada in Costa Rica. She wanted me to see if I could build a data logger that would capture the temp at different depths of the nest, and once they hatched they would sex the turtles. Ultimately I could not do it cost effectively, but I could nearly compete with the consumer products because they can be pricey for what they wanted to accomplish. If you are interested I can see if she can suggest some white papers.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates